In elmerfer/RAPInetMHCpan: an R API to netMHCpan and netMHCIIpan softwares for neural network prediction of peptide binders

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

This is an R Application Program Interface for the well known softwares netMHCpan and netMHCIIpan. The RAPInetMHCpan package allows easy deployment of such softwares into your own machine, facilitates their used trhrough full R code and allow multicore processing.

Installing netMHCpan and netMHCIIpan

The RAPInetMHCpan library allows you installing both softwares from their tar ball zipped files (up to now only version 4.0 for both softwares). NOTE: please red and follow the Readme from the main github page.

From Technical University of Denmark follow the instructions and fill the form to receive the rights to download netMHCpan and netMHCIIpan. Save them to your favorite directory ("/home/../myfavoritedir/..").

Open an R session or RStudio and type:

installNetMHCPan(file = "/home/.../myfavoritedir/netMHCpan-4.0a.Linux.tar.gz" , data = NULL, dir = "/where i whant/dir")
installNetMHCIIPan(file = "/home/.../myfavoritedir/netMHCIIpan-4.a.Linux.tar.gz" , data = NULL, dir = "/where i whant/dir")

It will print on console:

netMHCpan Installation OK

or

netMHCIIpan Installation OK

Running

library(seqinr)
library(stringr)
library(BiocParallel)
library(openxlsx)
library(RAPInetMHCpan)

#this sequence segment belongs to NA H5N1 .[(GenBank: AAC32089.1)
myseq <- s2c("MNPNQKIITIGSICMVVGIISLMLQIGNIISVWVSHIIQTWHPNQPEPCNQSINFYTEQAAASVTLAGNSSLCPISGWAIYSKDNSIRIGSKGDVF")

seqfile <- file.path(.libPaths()[1],"filetest.fasta")
write.fasta(myseq, file.out = seqfile,
            name = "NA H5N1 .[(GenBank: AAC32089.1) short sequence]")
if(!file.exists(seqfile)){
  stop("ERROR: pls check path")
}
##Run for a unique allele for MHC I
t <- Sys.time()
ret <- RunNetMHCPan(seqfile = seqfile, allele = "HLA-A01:01")
print(Sys.time()-t)
knitr::kable(ret, "pandoc")

These table shows the RAPIMHC result class objects with the.

The following plot shows the first Peptide and its interaction core surrounded by squares. The intermediate segment refers to a "deletion" interpreted as a peptide segment bulging out from the MHC binding cleft.

PlotPeptideCore(eDB = ret, index = 1)

The following function will display all the Peptides logo of different lengths. If several peptides of the same length are found, they will be displayed together.

PeptideBindersLogo(eDB = ret)

elmerfer/RAPInetMHCpan documentation built on Sept. 17, 2021, 2:14 p.m.